IB/mlx4: Eliminate duplicate barriers on weakly-ordered archs

Code includes wmb() followed by writel(). writel() already has a barrier on
some architectures like arm64.

This ends up CPU observing two barriers back to back before executing the
register write.

Since code already has an explicit barrier call, changing writel() to
writel_relaxed().

Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Reviewed-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
This commit is contained in:
Sinan Kaya 2018-03-19 22:47:44 -04:00 committed by Jason Gunthorpe
parent 185899ee8d
commit 97d82a48d7

View file

@ -3894,8 +3894,8 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
*/
wmb();
writel(qp->doorbell_qpn,
to_mdev(ibqp->device)->uar_map + MLX4_SEND_DOORBELL);
writel_relaxed(qp->doorbell_qpn,
to_mdev(ibqp->device)->uar_map + MLX4_SEND_DOORBELL);
/*
* Make sure doorbells don't leak out of SQ spinlock